Liu W, Liu M, Guo X, et al. - In order to take account of the deep learning algorithms to identify and determine the clot burden of acute pulmonary embolism (APE) on computed tomographic pulmonary angiography (CTPA), a retrospective study was performed to include a sum of 590 individuals (460 with APE and 130 without APE) who underwent CTPA. No statistically significant difference was found in the area under the curve (AUC) with the different probability thresholds. The sensitivity and specificity of U-Net in detecting clot respectively were 94.6% and 76.5% while the AUC was 0.926, when the probability threshold for segmentation was 0.1. Furthermore, this research revealed that the clot burden estimated with U-Net was significantly associated with the Qanadli score, Mastora score, and right ventricular functional parameters on CTPA. For the detection of pulmonary emboli, DL-CNN achieved a high AUC and can be used to quantitatively measure the clot burden of APE individuals, which may contribute to decreasing the workloads of clinicians.